This article aims to measure energy poverty in Colombia in its 32 departments and its capital city from 2018 to 2021, using a composite approach. To achieve this, a multidimensional energy poverty index (MEPI) was designed, according to the methodology proposed by Nussbaumer et al. (2012, 2013). Seventeen variables were used, which were distributed across 6 dimensions, and recorded by the National Quality of Life Survey (ECV, Spanish acronym), administered by the National Administrative Department of Statistics (DANE) of Colombia. In addition, a nested weighting method was used to assign weights within the index. Subjective weights were given to the dimensions, and an entropy method was used for each of the component variables. The results show that energy poverty has remained constant in Colombia across the period studied, both in the cities and in small towns and rural areas. There are significant differences between these areas in all territories and the departments located in the most remote areas of the country have a higher energy poverty. This is consistent with the low population density, as well as with areas that are not interconnected (ANI) with the national interconnected system (SIN, Spanish acronym).